The present invention generally relates to centrifuges and, more particularly, to centrifuges employed to remove particulates from lubricants.
Centrifuges have often been employed to remove various particulate contaminants from lubricating oil of internal combustion engines. The most common applications of centrifuges in this context have been in large diesel engines. Typically, lubricating oil of a large diesel engine may be continuously passed through a full flow filter and through a bypass centrifugal filter or centrifuge. While conventional centrifugal filters may be relatively costly, their cost is justified because engine life is improved when they are used.
Recent developments in environmental standards have introduced additional demands on filtering systems for diesel engine oil. Injector timing retardation is needed to meet more stringent air pollution standards. These demands result in increased production of carbon soot on the cylinder walls of an engine. Soot finds its way into the lubricating oil of the engine. Conventional full flow filters and conventional centrifugal filters do not adequately remove soot from the oil. Engine life is reduced in the presence of soot in the oil because the soot is abrasive and it reduces lubricating qualities of the oil.
Various efforts have been made to improve performance of centrifuges in attempts to introduce soot removal capabilities. Some examples of these efforts are illustrated in U.S. Pat. No. 6,019,717, issued Feb. 1, 2000 to P. K. Herman and U.S. Pat. No. 6,984,200 issued Jan. 10, 2006 to A. L. Samways. Each of these designs is directed to a problem of removing very small particles of soot, i.e., particles of about 1 to about 2 microns. Centrifuges separate particulates from fluids by exposing the particulates to centrifugal forces. Particulates with a density greater than the fluid are propelled radially outwardly through the fluid. But, in the case of soot particles suspended in oil, separation is difficult because soot particles have a density very close to oil. Consequently, very high centrifugal forces may be required to move the soot particles through oil. Typically centrifugal forces of about 10,000 g's may be needed. These high forces may be produced by rotating a centrifuge at very high speeds. Alternatively, the requisite high g forces may be produced within a centrifuge having a very large diameter. However, as a practical matter, it is desirable to limit the diameter of a centrifuge to diameter of about 7 to 10 inches to meet space limitation on a vehicle and to limit rotational inertial effects. Also there is a practical limitation on the rotational speed that can be imparted to a centrifuge. Speeds of about 10,000 to about 12,000 rpm represent the limits of the current state of the art.
In attempts to capture small soot particles within these practical speed and size parameters, prior art centrifuges employ complex and labyrinth-like oil passage pathways. As oil traverses these complex pathways, it remains in a centrifuge for a relatively long time. In other words, it has an extended “residence time”. It has heretofore been assumed that improved soot removal is directly related to increased residence time.
But, in various efforts to increase residence time, prior art centrifuges have employed oil passage pathways that introduce multiple changes in direction of flow of oil. Many of these changes in flow direction may be abrupt. As oil flow makes these abrupt changes in direction, vortices may be generated. These vortices may propagate throughout the entire mass of oil that may be present in a prior art centrifuge, resulting in oil flow that is turbulent in nature. Turbulence in oil flow may produce additional difficulty in removing small particles from the oil. Whenever any one particle is propelled outwardly by centrifugal force in a turbulent flow, there is a high probability that the particle will encounter a reverse flow of oil in a vortex. Such a reverse flow may propel the particle inwardly and thus cancel the desired effects of centrifugal force imparted by the centrifuge. Thus, the particle has a high probability of remaining suspended in the oil.
It can be seen that soot removal effectiveness of centrifuges in the present state of the art is bounded by various limiting conditions. First, there is a practical limit on a diameter of a centrifuge. Second, there is a practical limit on the rotational speed at which a centrifuge may be operated. And third, increased residence times may be attained at the cost of producing turbulent flow in a centrifuge. As described above, turbulent flow may offset or cancel any beneficial effects of increasing residence time. There has been no recognition in the prior art of a simple expedient to increase the soot removal effectiveness of centrifuges within the practical limits of centrifuge size and rotational speed.
As can be seen, there is a need for improvement of soot removal effectiveness in a practical centrifuge.